Ragas is the standard framework for measurable RAG evaluation: hallucination, faithfulness, relevancy

mdancho84 · x · 2026-09-19

The thread's #6 pick is Ragas, described as the standard framework for evaluating RAG pipelines — hallucination, faithfulness and relevancy are all measurable, and it can generate production-aligned test sets when you don't have one. The adjacent pick is Qdrant, a production vector database written in Rust (34.7k stars) that JDs list alongside Pinecone, Chroma, and FAISS.

Related event: Matt Dancho Lists 10 GitHub Repos to Become an AI Engineer in 90 Days(11 posts)→

Original post →

More from coding & agent

coding & agent channel →